Iranian EFL Learners ' Willingness to Communicate across Different Context-and Receiver-Types

Willingness to communicate is the most basic orientation toward communication. Almost anyone is likely to respond to a direct question, but many will not continue or initiate interaction. This paper investigates Iranian EFL learners' perceptions of their willingness to initiate communication across four types of context and three types of receiver. The study employed a questionnaire consisting of 20 situations in which a person might choose to communicate or not to communicate. The study concludes that learners were highly willing to communicate in two context-types (Group Discussion, & Meetings) and one receiver-type (Friend). They were not willing to initiate communication in other situations. The main reason is that majority of Iranians have the experience of communicating in English only in language classrooms in which they can have some group discussion, meetings, and friendly chat. They don't have an access to a native speaker or possibility to travel to an English speaking country. In general, it can be said that Iranian EFL learners are willing to initiate communication in situations experienced before, like group discussion or communicating with their friends. They don't feel confident enough to initiate communication in unfamiliar situations like public speaking. Therefore, contextand receiver-type familiarity is an effective factor for the situation in which a learner initiates communication.


Introduction
The current communicative approaches to second language (L2) instruction emphasize the importance of learners using the L2 in oral and written tasks.These approaches to instruction are based on the premise that learners' competence in the L2 is developed via performance and are supported by the dominant theories of second language acquisition (e.g., Long, 1996;Swain, 2000).This focus on the active use of the L2 in the language classroom has led to the emergence of an important construct in the field of L2 instruction: willingness to communicate (WTC).
Willingness to communicate is defined as a learner's ''readiness to enter into discourse at a particular time with a specific person or persons, using a L2" (MacIntyre et al., 1998, p. 547).As Do¨rnyei (2003) points out, competence in the L2 may not be enough.Learners need to be not only able to communicate but also willing to communicate in the L2.Research has shown that a learner's WTC influences how frequently the learner actively engages in communicating in the L2 (Cle´ment et al., 2003;Yashima et al., 2004).Thus MacIntyre et al. (1998) propose that WTC in L2 should be conceptualized as the primary goal of language instruction and as a comprehensive conceptual framework to describe, explain and predict L2 communication behavior.
Early models of WTC (e.g., MacIntyre, 1994) depicted WTC as being predicted by two variables: perceived communication competence and communication anxiety.That is, the model predicted that high levels of perceived competence combined with low levels of anxiety would lead to greater WTC and in turn more frequent communication in the L2.In 1998, MacIntyre et al. proposed a multi-layered pyramid model of WTC.The model differentiates between stable enduring influences (such as personality traits) and situation specific influences informing communication behaviour (e.g., desire to communicate with a specific person).Informed by the communication literature on willingness to communicate in the L1, the authors suggest that WTC depends on a range of factors such as the degree of acquaintance between communicators, the number of people present, the formality of the situation, and the topic of discussion.
In a more recent discussion of the model, MacIntyre (2007) emphasizes the complexity of the WTC construct, and the need to clearly define the variables investigated.For example, communication anxiety can operate at the individual level (i.e., an anxious person), or be triggered by the situational contexts (e.g., the language class) or by a particular event.The distinction is important not only in terms of measuring and understanding how these variables affect WTC but also in terms of pedagogical intervention.Anxiety triggered by situational factors is more likely to fluctuate over time and is perhaps more amenable to instructional intervention.Furthermore, MacIntyre calls on researchers to use methodologies which can capture the dynamic nature of this construct.The current study set out to investigate Iranian EFL learners' willingness to initiate communication across different context-and receiver-types.In other words, the researcher wants to probe if the learners' willingness to initiate communication varies in relation to different types on context and receiver (the person you are talking to).In order to capture the learners' perceptions, a self-assessment questionnaire was used.
The research questions guiding this study were: Q1-What were the Iranian EFL learners' perceptions regarding their willingness to initiate communication across different context-types?Q2-What were the Iranian EFL learners' perceptions regarding their willingness to initiate communication across different receiver-type?

Related Empirical Studies on WTC
Studies on WTC, like early studies investigating learners' motivation to study a L2, have tended to rely on data collected at one point of time, often collected via a single instrument, and to consider only quantitative findings.For example, the large scale cross-sectional study by MacIntyre et al. (2002), which investigated the effects of age and gender on WTC, used a questionnaire which asked the participants to rate themselves on eight scales.This questionnaire was administered to the participants only once.
More recent studies have attempted to address this gap in research on WTC by collecting data from a range of sources and over a period of time.For example, the study by Cao and Philp (2006) used a number of data collection tools: student questionnaires (self reports on trait level variables), eight classroom recordings and observations (over one month), and interviews with individual learners conducted at the end of the study to elicit information about situational variables.However, the small scale nature of the study (n = 8) precluded the researchers from providing clear correlations between the learner reported individual and situational factors and observed classroom behaviour.The four factors which were most frequently mentioned by the participants as affecting WTC included group size, self-confidence, and familiarity with the interlocutors.These findings supported the findings reported by another small scale study conducted by Kang (2005).Kang (2005) used a similar range of data collection tools but the number of participants was even smaller (n = 4).Kang identified three variables that contributed to the participants' WTC: security, excitement, and a sense of responsibility.Each of these variables was further affected by factors such as topic, interlocutor and conversational context.The notion of security is perhaps similar to that of anxiety and was shaped by relative familiarity among the interlocutors, the size of the group, and the L2 fluency of fellow learners.All the participants reported that they felt less secure and hence less willing to communicate when they perceived other group members to be more fluent than they were.Thus relative L2 proficiency in relation to other learners' perceived proficiency was identified as an important factor in WTC.Other factors affecting feelings of security or confidence included topic familiarity and interest.A greater level of familiarity and interest in the topic lowered feelings of insecurity or raised excitement and hence learners expressed a greater WTC.Although the study led to a proposed preliminary model of situational WTC in L2, the model is static.It is not clear whether learners' perception of these situational variables changed over the course of the semester.
Another interesting finding in Kang's study relates to the learners' attitudes to working in groups with fellow L1 (Korean) speakers.This is particularly pertinent, given the emphasis on small group work advocated by communicative approaches to L2 instruction.Kang reports that learners commented about their reluctance to speak in the L2 in groups composed of fellow L1 speakers.One participant commented about this 'unnatural' situation: ''I feel like I'm wearing a mask" (p.284).
The importance of learners' attitudes towards classroom tasks and activities has been highlighted in recent research and theoretical discussions on second language learning motivation.For example, whereas earlier models of motivation focused on forces operating at the macro level (e.g., integrative or instrumental orientation); the more recent process model (Do¨rnyei and Otto, 1998) focuses on situation specific forces which can better explain learner behaviour observed in the classroom.The study conducted by Do¨rnyei and Kormos (2002) in British and Hungarian classes found that students' engagement in classroom oral activities (measured by number of words and turns) correlated significantly with attitudes towards the language tasks they were asked to perform.
Learners' attitude to classroom activities is also emphasized by researchers working from a sociocultural theoretical perspective.From this perspective, learners are viewed as active agents (Lantolf and Thorne, 2006) who assign relevance and significance to things and events in their life.This notion of learners as active agents in their learning is particularly apt when investigating adult classroom contexts.Such learners often display a wide range of learning goals, expectations and abilities.
To summarize, recent theoretical discussions of the construct WTC have emphasized the need to collect data that capture the dynamic nature of this construct.However, research to date, which has attempted to do so, has tended to be small scale, and although data were collected over a period of time, there is very little mention of whether students' perceptions changed over time.Furthermore, although attitudes to the tasks students are asked to perform have been noted as an important consideration in explaining students' willingness to actively contribute to the task, there has been relatively little classroom-based research on learners' attitude towards such tasks.

Study
The study was conducted in a "Panel Discussion" class, at a private English language institute (Arses), Amol.Panel Discussion is the most advanced course available for EFL learners in the institute.Thus students entering Panel Discussion are typically those who had been learning English for 4 years in the institute: 2 years at introductory levels, 1 year at intermediate level, and 1 year at upper intermediate level.
Learners in Panel Discussion class are taught entirely in English.Class activities are based on themes (e.g., identity and stereotypes; cross-cultural communication) which are explored through two novels that learners are required to read and through text and non-text materials (e.g., television shows and blogs).The teacher attempts to link the selected theme with the learners' own knowledge of the world whenever possible, to ensure relevance and engagement with topics and tasks.
Typically, learners are expected to read (e.g., a few pages of the novel and a journal article) and/or complete a task for homework.In class, (25 learners), students work in groups of three or four to discuss matters arising from the text or task, including linguistic difficulties, interesting topical issues etc.Findings are then reported to other class members in a whole class discussion.As the class discussion unfolds, the teacher facilitates the exchange of ideas and opinions by providing learners with the linguistic tools necessary to sustain the discussion (i.e., adequate vocabulary and sentence construction).
Assessment is based on a number of tasks.Oral tasks contribute to 45% of the grade.The assessment tasks include two tasks which involve some prior preparation: a formal, researched pair presentation (25%) and two group debates (10%) on a topic related to the themes covered in the class.The remaining 10% is allocated to class participation.Class participation was defined to students as ''your level of input in English in class discussion, small group discussions and other class interaction regardless of your proficiency level in English".In other words, the emphasis is on fluency not on accuracy.
Self-assessment (SA) questionnaire was introduced in Panel Discussion not as an alternative form of assessment but as a means of encouraging learners to become more reflective and autonomous learners.As Boud (1995) and Oscarson (1989), among others, argue self-assessment is a valuable pedagogical activity per se.In this context it was introduced in order to encourage learners to reflect on their own willingness to initiate communication, identify their strengths and weaknesses in relation to their oral proficiency and set goals accordingly for the semester.The reasons for introducing SA questionnaires were communicated to the students at the beginning of the semester.Learners were asked to complete SA questionnaires in class.
The self-assessment questionnaire measures a person's willingness to initiate communication.It contained 20 situations in which a person might choose to communicate or not to communicate.The learners had completely free choice.They were supposed to indicate the percentage of times they would choose to communicate in each type of situation.In other words, the students indicated in the space at the left of the item what percent of the time they would choose to communicate (0=Never to 100=Always).The face validity of the instrument is strong, and the results of extensive research indicate the predictive validity of the instrument.Alpha reliability estimates for this instrument have ranged from .85 to well above .90.Of the 20 items on the instrument, 8 are used to distract attention from the scored items.The twelve remain items generate a total score, 4 context-type scores, and 3 receiver-type scores.The sub-scores generate lower reliability estimates, but generally high enough to be used in research studies.(See Appendix A).
To compute the total WTC score, the researchers added the average scores for stranger, acquaintance, and friend.Then they divided it by 3.All scores, total and sub-scores, will fall in the range of 0 to 100.There were also some norms for WTC scores in terms of the following features:

Findings
The students' perceived willingness to initiate communication is presented in table 1.As table 1 shows, the learners are highly willing to communicate in two context-types (Group Discussion: 93.80, Meeting: 92.52), and also in one friend-type (Friend: 95.40).In other sub-groups, the learners' willingness to communicate was low (Interpersonal Conversation: 54.04,Public Speaking: 26.96,Stranger: 13.16,Acquaintance: 44.96,Friend: 95.40).All this information is illustrated graphically in Graph 1.

Discussion
A new perspective that there are situational variables that have the potential to change an individuals' WTC has challenged the perspective that views WTC as a trait-like predisposition.This new perspective was well established in MacIntyre et al. (1998), in which L2 WTC was defined as ''a readiness to enter into discourse at a particular time with a specific person or persons, using a L2'' (p.547).According to their heuristic model, WTC is influenced by immediate situational antecedents -the desire to communicate with a specific person and the state of communicative self-confidence -and more enduring influences, such as interpersonal motivation, intergroup motivation, self-confidence, intergroup attitudes, social situation, communicative competence, intergroup climate, and personality.
Reflecting this situational view, researchers have investigated how WTC can be affected by situational variables, such as social contextual variables.MacIntyre et al. (2001) showed that social support, particularly from friends, influenced WTC outside the classroom.In Baker and MacIntyre's (2000) study comparing French immersion versus non-immersion programs, it was found that WTC was affected by the learning context.Cle´ment et al. (2003) also showed that L2 WTC was influenced by the frequency and quality of L2 contact, which interacted with each other, through the mediation of L2 confidence.
These findings, as well as MacIntyre et al._s (1998) model, have convinced us that WTC is subject to situational variables.These previous studies, however, examined situational variables mainly through a quantitative method using questionnaires, which is not insightful enough to explore situational characteristics of WTC in an actual situation.
The study concludes that learners were highly willing to communicate in two context-types (Group Discussion, & Meetings) and one receiver-type (Friend).They were not willing to initiate communication in other situations.The main reason is that majority of Iranians have the experience of communicating in English only in language classrooms in which they can have some group discussion, meetings, and friendly chat.They don't have an access to a native speaker or possibility to travel to an English speaking country.In general, it can be said that Iranian EFL learners are willing to initiate communication in situations experienced before, like group discussion or communicating with their friends.They don't feel confident enough to initiate communication in unfamiliar situations like public speaking.Therefore, context-and receiver-type familiarity is an effective factor for the situation in which a learner initiates communication.The learners seem to feel more secure with familiar contexts and situations.In other words, they feel safe from the fears that non-native speakers tend to have in L2 communication in familiar situations.Security mainly seemed to be determined by interlocutors.An interlocutor who had not had English conversations with the participants and did not know their English proficiency tended to make them feel insecure.That is why the learners were not willing to initiate communication with strangers and the people who were not their friends.One reason of this feeling is that, they may be afraid of making mistakes.Actually, security was shaped by the relative familiarity among the interlocutors, which can influence power dynamics among them.In this regard, this study is in congruence with the studies mentioned above in that both familiarity with context-type and receiver-type affects the EFL learners' willingness to communicate.

Pedagogical implications
Based on the findings, we propose some suggestions that can contribute to generating L2 learners' situational WTC in ESL classes and conversation partner programs.To generate situational WTC, topics in which L2 learners are interested, about which they have background knowledge, with which they have experience, and which can drive their personal or intergroup motives need to be offered.Given that L2 learners' excitement can decrease if they talk about the same topic over and over again, we also recommend that different topics be discussed both within a lesson and across lessons to create L2 learners' situational WTC.
Brainstorming, taking a survey, and letting L2 learners bring in topics are good ways to identify both the most commonly occurring interests and shared background knowledge of the entire class, and those of individual L2 learners.At this point, one question arises: How can I deal with various L2 learners who want to talk about different topics?Arranging the discussion group based on their own selected topics, accommodating different topic preferences among L2 learners, is the best way to generate situational WTC.Topics that are identified as general interests can be presented as discussion topics in order to create situational WTC for the greatest number of class members.
Efforts to create a safe environment, in which students do not have a high level of fear about making mistakes or producing errors, should be made, by listening to them carefully, smiling and providing some active responses.These efforts should be made especially in the beginning, when L2 learners' insecurity is maximized, and especially for those with a lower level of proficiency, who tend to have a high level of fear of producing incorrect utterances and losing face.
We also recommend that the social background of L2 learners as well as their proficiency be considered in class assignments and discussion group assignments in order to prevent L2 learners from decreasing their excitement and security in a class or a discussion group where many students are of the same community.Decreasing the number of L2 learners in a group can also contribute to creating situational WTC by increasing responsibility and security.
Most importantly, based on the fact that when more facilitating factors are provided, a greater degree of WTC can be created, ESL teachers or conversation partners should try to provide the factors facilitating WTC as much as possible, instead of focusing on one factor at the expense of other facilitating factors.In so doing, we can generate or increase L2 learners' situational WTC, which can contribute to their successful SLA.

Table 1 .
Descriptive statistics variables in the study